An Alternative Purification Method for Human Serum Paraoxonase 1 and its Interactions with Sulfonamides

Authors

  • Deniz Ekinci,

    1. Atatürk University, Science Faculty, Department of Chemistry, Biochemistry Division, 25240, Erzurum, Turkey
    2. Ondokuz Mayıs University, Agricultural Faculty, Department of Agricultural Biotechnology, Enzyme and Microbial Biotechnology Division, 55200, Samsun, Turkey
    Search for more papers by this author
  • Murat Şentürk,

    1. Atatürk University, Science Faculty, Department of Chemistry, Biochemistry Division, 25240, Erzurum, Turkey
    2. Ağrıİbrahim Çeçen University, Science and Art Faculty, Chemistry Department, 04100 Ağrı, Turkey
    Search for more papers by this author
  • Şükrü Beydemir,

    Corresponding author
    1. Atatürk University, Science Faculty, Department of Chemistry, Biochemistry Division, 25240, Erzurum, Turkey
    2. Atatürk University, Biotechnology Application and Research Center 25240, Erzurum, Turkey
    Search for more papers by this author
  • Ömar İrfan Küfrevioğlu,

    1. Atatürk University, Science Faculty, Department of Chemistry, Biochemistry Division, 25240, Erzurum, Turkey
    Search for more papers by this author
  • Claudiu T. Supuran

    Corresponding author
    1. Università degli Studi di Firenze, Polo Scientifico, Laboratorio di Chimica Bioinorganica, Rm. 188, Via della Lastruccia 3, 50019 Sesto Fiorentino (Florence), Italy
    Search for more papers by this author

Corresponding authors: Şükrü Beydemir, beydemir@atauni.edu.tr; Claudiu T. Supuran, claudiu.supuran@unifi.it

Abstract

Paraoxonase 1 (PON1), a high-density lipoprotein (HDL)-associated esterase, is known to mediate antioxidant and antiatherogenic properties. Purification of PON1 has been challenging for a long time. Here, we report a novel purification technique for this enzyme, which allowed us to obtain human serum paraoxonase 1 (hPON1) using straightforward chromatographic methods, such as Diethylaminoethyl-Sephadex anion exchange chromatography and Sepharose 4B–4-phenylazo-2-naphthaleneamine hydrophobic interaction chromatography. We purified the enzyme 302-fold with a final specific activity of 4775 U/mg and a yield of 32%. Furthermore, we examined the in vitro effects of some sulfonamide derivatives, such as sulfacetamide, homosulfanilamide (mafenide), sulfosalazine, furosemide, acetazolamide, and 1,3,4-thiadiazole-2-sulfonamide on the enzyme activity to better understand the inhibitory properties of the molecules. The six sulfonamides dose-dependently decreased the activity of hPON1 with inhibition constants in the millimolar – micromolar range. This study provides an efficient method, which may be useful for other enzymes such as those related to acetylcholinesterase. It also demonstrates the off-target activity of sulfonamides.

Abbreviations:
DEAE

Diethylaminoethyl

HDL

high-density lipoprotein

hPON1

Human serum paraoxonase 1

LDL

low-density lipoprotein

OP

Organophosphate

PON1

Paraoxonase 1

SDS

Sodium dodecyl sulfate

Human serum paraoxonase/arylesterase 1 (EC 3.1.8.1, hPON1) is exclusively associated with high-density lipoprotein (HDL) and is a genetically polymorphic enzyme that has the ability to hydrolyze a wide range of organophosphates (OPs) and carboxyesters (1). This calcium enzyme is synthesized in the liver and then secreted into the blood. Its gene family contains at least three members in mammals: PON1, PON2, and PON3 (2). PON1 and PON3 are expressed primarily in the liver, while PON2 is widely expressed in various tissues, including brain, liver, and kidney (3–5). The enzyme active site contains a Ca(II) ion critical for catalysis (and a second structural such ion) co-ordinated by five protein residues (the side chain oxygens of Asn224, Asn270, Asn168, Asp269, and Glu53) and by a water molecule which in deprotonated form (as hydroxide ion) acts as nucleophile for the hydrolysis of substrates (1). Another potential ligand of the catalytic metal ion is an oxygen of the phosphate substrate that will undergo hydrolysis, but the rate determining step of the catalytic turnover is the deprotonation of the calcium-co-ordinated water molecule, assisted by a His–His dyad (1). This is in fact reminiscent of some zinc enzymes, such as the carbonic anhydrases (CAs, EC 4.2.1.1) in which a Zn(II) hydroxide species is the catalytically active nucleophile, which is generated by deprotonation of zinc-co-ordinated water assisted by a His residue from the enzyme active site(6).

High-density lipoprotein particles are involved in the protection of low-density lipoprotein (LDL) against oxidative modification. This mechanism is related to the activity of the Paraoxonase (PON) enzymes, which are located on HDL (7). Thus, PON1 is thought to play an antiatherogenic role by inhibiting the oxidation of LDL and hydrolyzing lipid peroxides (8). The enzyme is distinctive in that it interacts with serum lipoproteins. This association, in combination with studies showing that PON1 can protect LDL from oxidative damage in artificial systems, suggests that PON1 has a protective role in vivo and inhibits the development of vascular and coronary disease (9).

Paraoxonase is important in the metabolism also as an organophosphate (OP) hydrolyzer, and it hydrolyzes aromatic carboxylic esters, such as phenyl acetate, being thus involved in drugs and xenobiotics metabolism. Moreover, it hydrolyzes various lactones, including naturally occurring lactone metabolites (10,11). Paraoxon (diethyl 4-nitrophenyl phosphate) is a parasympathomimetic, which acts as an acetylcholinesterase inhibitor. It is an OP oxon and the active metabolite of the insecticide parathion. It is also used as an opthalmological drug against glaucoma. Paraoxon is one of the most potent acetylcholinesterase-inhibiting insecticides available, around 70% as potent as the nerve agent sarin and so is now rarely used as an insecticide (12) because of the risk of poisoning to humans and other animals (Figure 1).

Figure 1.

 Molecular structure of paraoxon.

Paraoxonase has been purified so far from different sources with different yields and purification folds (13–17). However, most of these previous studies either included many steps or had low purification yields. For these reasons, we focused in this study to develop a new and simpler chromatographic method for the purification of hPON1. There are slight differences among the molecular weigths of paraoxonase enzymes from various sources. The human PON1 has a molecular weight of 43–45 kDa (18), rabbit liver paraoxonase 1: 40 kDa (19), human PON2: 37 kDa, rabbit serum PON3: 40 kDa (20), rat liver PON1: 43 kDa (21) [see Fig 7].

Figure 7.

 Molecular structures of the sulfonamide derivatives investigated as human serum paraoxonase 1 inhibitors.

As sulfonamides constitute an important class of drugs, having several types of pharmacological features, such as antibacterial (22), antitumor (23), anti-CA (6,23–26), diuretic (27,28), and protease inhibitory activity (29–32), we decided to investigate whether such compounds interfere with hPON1 activity. Some sulfonamides are also the strongest inhibitors of various enzymes such as the CAs (6,23,33). The sulfonamides investigated in this study, of types 1-6 (sulfacetamide, mafenide, sulfosalazine, furosemide, acetazolamide, and 1,3,4-thiadiazole-2-sulfonamide), are commonly clinically used antibiotics or CA inhibitors (31–33). There is no information in the literature regarding the effects of sulfonamides on hPON1 activity.

Experimental

Materials used in this work, including Diethylaminoethyl (DEAE)-Sephadex A50, Paraoxon, protein assay reagents, and chemicals for electrophoresis, were obtained from Sigma-Aldrich Chemie 82024 Taufkirchen, Germany. All other chemicals used were of analytical grade and obtained from either Sigma-Aldrich or Merck KGaA Darmstadt, Germany.

Paraoxonase activity assay

Paraoxonase activity was determined at 25 °C with paraoxon (diethyl 4-nitrophenyl phosphate) (1 mm) in 50 mm glycine/NaOH (pH 10.5) containing 1 mm CaCl2. The enzyme assay was based on the estimation of p-nitrophenolate formation, which has been followed spectrophotometrically at 412 nm. The molar extinction coefficient of p-nitrophenoxide (ε = 18 290/m per cm at pH 10.5) was used to calculate the enzyme activity (34). One enzyme unit was defined as the amount of enzyme that catalyzes the hydrolysis of 1 μmol of substrate at 25 °C. Assays were performed using a CHEBIOS UV-VIS spectrophotometer.

Ammonium sulfate precipitation

Fifteen milliliters of Triton X-100–treated human serum was precipitated with ammonium sulfate. The precipitation intervals for paraoxonase were 60–80% (16). The precipitate was collected by centrifugation at 1500 g for 20 min and redissolved in 100 mm Na-phosphate buffer (pH 7.0).

DEAE-Sephadex A50 anion exchange chromatography

The enzyme solution, which had been dialyzed in the presence of 1 mm Na-phosphate buffer (pH 7.0) at 4 °C, was loaded onto the DEAE-Sephadex A50 anion exchange column (3 cm2 × 30 cm), which had been equilibrated with 100 mm Na-phosphate buffer (pH 7.0). The column was washed with 100 mm Na-phosphate buffer (pH 7.0), and then elution was performed with a linear gradient of 0–1.5 m NaCl (35–36). Eluted fractions were collected, and enzyme activity was checked at 412 nm. Tubes with enzyme activity were combined. All purification procedures were performed at 4 °C.

Synthesis of Sepharose 4B–4-phenylazo-2-naphthaleneamine hydrophobic interaction gel

Three gram sepharose 4B was added in 50 mL water after washing, and 0.05 g KMnO4 was added in this solution. The mixture was stirred for 2 h, was taken in a buchner funnel, and washed with, 200 mL water, 100 mL ethanol, and 50 mL acetone, respectively. The mixture was taken in 20 mL acetone and was added 20 μL SOCl2. It was then taken in the buchner funnel and washed with 50 mL acetone; 50 μL aniline was added and stirred for 30 seconds. 40 mg 2-naphthaleneamine was dissolved in 1 m HCl. 120 mg NaNO2 was dissolved in 10 mL water and added in that solution at 4 °C. After adjusting the pH to 10, the reaction was completed in 45 min. After washing with 300 mL 100 mm phosphate buffer pH 7 including NaCl, the gel was loaded onto the column.

Preparation of hydrophobic interaction chromatography column

Fractions from the DEAE-Sephadex column were loaded onto the Sepharose 4B–4-phenylazo-2-naphthaleneamine hydrophobic interaction column, which had been equilibrated with 100 mm Na-phosphate buffer (pH 7.0). Elution was performed using NaCl gradient. Fractions were analyzed for both protein amount (280 nm) and enzyme activity (412 nm), as shown in Figure 2. Tubes with enzyme activity were combined for other kinetic studies.

Figure 2.

 Purification of human serum PON1 by Sepharose 4B–4-phenylazo-2-naphthaleneamine hydrophobic interaction chromatography with NaCl gradient. Qualitative protein concentration was determined by measuring an absorbance at 280 nm and human serum paraoxonase 1 activities of fractions were assayed using a paraoxon substrate. Other hydrophobic proteins were removed by measuring the enzyme activity.

Protein assay

During the purification steps, protein quantity was determined spectrophotometrically at 595 nm according to the Bradford method using bovine serum albumin as the standard (37–39).

SDS polyacrylamide gel electrophoresis

Sodium dodecyl sulfate polyacrylamide gel electrophoresis was performed after purification of the enzyme. It was performed with 10% and 3% acrylamide concentrations for the separating and stacking gels, respectively, and 0.1% SDS (32). Sample (20 μg) was applied to the electrophoresis medium. Gels were stained for 1.5 h in 0.1% Coomassie Brilliant Blue R-250, 50% methanol, and 10% acetic acid then destained with several changes of the same solvent without dye. The electrophoretic pattern is shown in Figure 3.

Figure 3.

 SDS–PAGE analysis of purified human serum paraoxonase 1. Lane 1: standard proteins (kDa): MBP-CBD (62), aldolase (47.5), triosephosphate isomerase (32.5), myoglobin (20.5). chicken egg white lysosime (16.5). Lane 2: purified human serum paraoxonase 1.

In vitro inhibition assay

We examined the inhibitory effects of six sulfonamide derivatives: furosemide, sulfosalazine, acetazolamide, sulfacetamide, mafenide, and 1,3,4-thiadiazole-2-sulfonamide (compounds 1-6) against the purified enzyme hPON1. All compounds were tested in triplicate at each concentration used. PON activities were measured in the presence of different sulfonamide derivative concentrations. Control activity was assumed to be 100% in the absence of inhibitor. For each substance, a percent activity versus drug concentration graph was drawn. For determination of Ki values, three different inhibitor concentrations were tested for each drug. In these experiments, paraoxon was used as substrate at five different concentrations (0.15, 0.3, 0.45, 0.6, and 0.75 mm). Lineweaver–Burk curves were used for the determination of Ki and inhibitor type (40–42).

Results and Discussion

Human serum paraoxonase 1 contains 354 amino acids, and its molecular weight is of 43 kDa (1,35). It is a HDL-associated enzyme and is evolutionarily rather conserved in vertebrates (except fish, birds, and invertebrates) (1,45). Organophosphates are substrates of PON1 and thus, this enzyme protects the nervous system against OP toxicity. It is called paraoxonase because of the fact that it hydrolyzes paraoxon (diethyl 4-nitrophenyl phosphate), a metabolite of the insecticide parathion. PON1 is also able to metabolize lipid peroxides and protect against LDL accumulation (3). It is suggested that hPON1 activity is responsible for the antioxidant effects on HDL (43).

Although PON has been purified from a wide range of sources using different purification procedures, but with many diverse steps being involved and thus the yields and purity of the enzyme were rather low (14–17). For example, Furlong et al. purified hPON1 with approx. 62.1- fold purification, using Agarose blue, Sephadex G 200, DEAE-Trisacryl M, Sepladex G 75 chromatography techniques (14). In another study, rat liver PON3 enzyme was purified 177-fold with a yield of 0.4% using six steps, including hydroxyapatite adsorption, DEAE-Sepharose CL-6B chromatography, Cibacron Blue 3GA non-specific affinity chromatography, anion exchange on Mono Q HR 5/5, DEAE-cellulose, and a final affinity chromatography on Concovalin A-Sepharose (15). In a previous study, we isolated PON1 237-fold from human serum with a yield of 51% and a specific activity of 3912.4 U × per mg using ammonium sulfate precipitation (60–80%), DEAE-Sephadex anion exchange, and Sephadex G-200 gel filtration chromatography (17).

A novel and efficient method was recently reported by Sinan et al. (16), who purified PON1 227-fold from human serum with a final specific activity of 1730.45 U × per mg and a yield of 72.54% using ammonium sulfate precipitation and CNBr-activated Sepharose 4B-L-tyrosine-1-naphytylamine hydrophobic interaction chromatography (16). In this study, we modified this method to obtain a cheaper and risk-free gel. We used ammonium sulfate fractionation (60–80%) (16), DEAE-Sephadex anion exchange, and Sepharose 4B–4-phenylazo-2-naphthaleneamine hydrophobic interaction chromatography for purifying hPON1. We purified the enzyme 302-fold with specific activity of 4775 U × per mg and a yield of 32%. In the hydrophobic interaction gel, we did not activate Sepharose 4B with CNBr, a highly toxic and dangerous compound, but we propose a novel activation method. Therefore, we obtained a risk-free and more economical gel, as CNBr-activated Sepharose 4B is also quite expensive. Moreover, we used aniline as spacer arm instead of tyrosine. In Sepharose 4B-L-tyrosine-1-naphytylamine gel, the functional -OH group is free on the tyrosine molecule, and this brings a more hydrophilic character to the gel. As for our gel, the functional group of aniline is an -NH2 group, which makes a covalent bond with Sepharose 4B. The summary of this new purification procedure is shown in Table 1. Thus, there are no free functional groups in our new gel, which is depicted schematically in Figure 4. For this reason, our gel has a more hydrophobic character. Thus, we thought that using aniline instead of tyrosine is an advantage for this kind of chromatographic purification methods (Figure 4).

Table 1.   Summary of the purification procedure of human serum paraoxonase 1
Purification stepActivity (EU/mL)Total volume (mL)Total protein (mg)Total activity (EU)Specific activity (EU/mg)Yield (%)Purification factor
Triton X-100–treated serum12615120189615.81001
Ammonium sulfate precipitation1001273.2120116.4631.04
Ion exchange chromatography84.391.277585984037.8
Hydrophobic interaction76.480.13611477532302
Figure 4.

 Schematic representation of the Sepharose 4B–4-phenylazo-2-naphthaleneamine hydrophobic gel. The functional group of aniline (−NH2) was covalently bound with Sepharose 4B by means of an amide bond. Sepharose-4B was oxidized by KMnO4 and subsequently activated by SOCl2. After that, aniline was attached to the activated gel as a spacer arm, and finally diazotized naphytylamine was clamped to the para position of aniline molecule as ligand. In this way, Sepharose 4B–4-phenylazo-2-naphthaleneamine hydrophobic interaction gel was obtained. The chromatography column was equilibrated with 100 mm phosphate buffer pH 7, including NaCl.

Many chemical species influence metabolism at low concentrations by decreasing or increasing the normal enzyme activity, especially by inhibiting enzymes with critical function (41), being thus drug targets (45). PON is important in the metabolism as OP hydrolyzer. Organophosphates (OP) are pesticides that inhibit cholinesterase. They cause poisonings and deaths (46–48). Paraoxonase, a member of the A-oxonase family, breaks down acetylcholinesterase inhibitors before they bind to the cholinesterases, and thus protects people from harmful effects caused by exposure to low doses of OP pesticides (44,45). Yet, it is estimated that worldwide, 220 000 people are killed each year from such exposures. This is one reason why inhibitors of paraoxonase must be well investigated. PON is also a drug target (49–52). We have performed a number of studies regarding the interactions of different inhibitors with several such enzymes, including PON1 (50–52). For instance, we examined the in vitro effects of some analgesic drugs, such as lornoxicam, indomethacin, tenoxicam, diclofenac sodium salt, ketoprofen, and lincomycine, on human serum PON1 activity and observed that these analgesics inhibited human serum PON1 at very low concentrations. Inhibition range of the drugs was as follows: lornoxicam > indomethacin > tenoxicame > diclofenac sodium salt > ketoprofen > lincomycine (31). Additionally, Tomás et al. (53) investigated the effect of the lipid-lowering drug simvastatin on serum PON activity in patients with familial hypercholesterolemia (FH). They found that serum PON1 activity toward paraoxon significantly increased during treatment with simvastatin. So, they hypothesized that simvastatin might have relevant antioxidant properties (53). Besides, Sinan and colleagues showed that gentamycin sulfate and cefazolin sodium salt dose- and time-dependently inhibited human serum PON1, with IC50 values of 0.887 and 0.0084 mm, respectively, but did not affect liver PON1 activity in human hepatoma HepG2 cells (16).

However, it is thought that more extensive inhibition studies are necessary for a better understanding of the protective role of PONs against the toxic effects of xenobiotics, including environmental heavy metals and oxidative stress by-products (15,16). But, there are only few studies regarding effects of drugs on PON1 activity in literature. Considering these, we report in the present study the effects of sulfonamides 1-6 against purified hPON1.

We investigated the in vitro effects of sulfacetamide 1, mafenide 2, sulfosalazine 3, furosemide 4, acetazolamide 5, and 1,3,4-thiadiazole-2-sulfonamide 6 on hPON1 activity (Table 2). The results showed that these compounds are inhibitors of PON1 enzyme, and thus, they must be used carefully, especially on the patients having diseases in which PON1 activity plays a role.

Table 2.   Inhibition data for compounds 1-6 against purified human serum paraoxonase 1
CompoundIC50 (mm)Ki (mm)Inhibition type
Sulfacetamide 10.18 ± 0.0020.086 ± 0.008Competitive
Mafenide 20.19 ± 0.0120.34 ± 0.047Non-competitive
Sulfosalazine 30.26 ± 0.0170.57 ± 0.081Non-competitive
Furosemide 40.36 ± 0.0380.702 ± 0.064Non-competitive
Acetazolamide 50.80 ± 0.0510.743 ± 0.092Competitive
1,3,4-Thiadiazole- 2-sulfonamide 61.24 ± 0.0652.01 ± 0.103Non-competitive

The sulfonamide derivatives were determined to be inhibitors with the following rank order of (1) > (2) > (3) > (4) > (5) > (6). The IC50 values for the compounds (1), (2), (3), (4), (5), and (6) were as follows: 0.18, 0.19, 0.26, 0.36, 0.8, and 1.24 mm, respectively (Table 2). Ki graphs show that sulfacetamide and asetazolamide inhibit the enzyme in a competitive manner with paraoxon as substrate (Table 2, Figures 5 and 6). It is known that paraoxon and p-nitrophenylacetate are substrates for PON enzyme. Sulfacetamide and asetazolamide include amide groups, and this may be the reason why they inhibit the enzyme in a competitive manner with the above-mentioned ester as substrate. However, this hypothesis should be verified experimentally, by means of X-ray crystallographic studies. The other compounds investigated here inhibited hPON1 in a non-competitive manner with paraoxon as substrate (Table 2, Figures 5–7). The only quite effective hPON1 inhibitor detected here is sulfacetamide, which with a KI of 86 μm may constitute an interesting lead for obtaining more effective such compounds.

Figure 5.

 Lineweaver–Burk graphs for purified human serum paraoxonase 1 activity from human serum at five different paraoxon substrate concentrations and at three different sulfacetamide (A), mafenide (B), and sulfosalazine (C) concentrations (0.22–0.28 mm).

Figure 6.

 Lineweaver–Burk plots for purified human serum paraoxonase 1 activity at five different paraoxon substrate concentrations and at three different furosemide (A), acetazolamide (B), and 1,3,4-thiadiazole-2-sulfonamide (C) concentrations (0.33–1.35 mm).

Acknowledgment

We are grateful to Prof. Oktay Arslan, Department of Chemistry, Balikesir University for helpful discussions.

Ancillary